This service is a simple example that shows a Docker
HEALTHCHECK
and a test for inducing an unhealthy state.
make config
choose the value for TIMEOUT in seconds. If you set this to a value
less than 30, the service will be considered healthy, otherwise
unhealthy.
make install
With the example service now running, watch the status:
watch make status
Check the output of the status for the current health. It should
initially say starting until the healthcheck either succeeds or
fails. The healthcheck requires the timeout to occur up to three
times, with 30s in between checks; so after ~2mins it should change to
either healthy or unhealthy.
The service has the label autoheal=true, so if you are running the
autoheal service, the unhealthy service should be
automatically restarted (otherwise the service will run forever in the
unhealthy state).